Performance Analysis Support for Object Oriented Parellel Scientific Applications

Performance Analysis Support for Object Oriented Parellel Scientific Applications Jeffrey Nesheiwat Boleslaw K. Szymanski The complex and computationally demanding nature of scientific applications has fueled research in the area of parallel computing. Moving from conventional uniprocessor systems to multiprocessor systems makes designing, developing, testing, tuning, and maintaining scientific codes much more difficult. These difficulties are outweighed by the signi?cant speedup that parallel computing can provide. Since the primary reason for writing parallel codes is speed [13], it comes as no surprise that performance analysis is a vital part of the development process. Analysis tries to determine if a given algorithm is as fast as it can be, where the program can be further optimized, and how efficiently the underlying system is being used. Department of Computer Science, Rensselaer Polytechnic Institute, Troy, NY 12/05/1998 cs-98-05

Performance Analysis Support for Object Oriented Parellel Scientific Applications

Jeffrey Nesheiwat

Boleslaw K. Szymanski

The complex and computationally demanding nature of scientific applications has fueled research in the area of parallel computing. Moving from conventional uniprocessor systems to multiprocessor systems makes designing, developing, testing, tuning, and maintaining scientific codes much more difficult. These difficulties are outweighed by the signi?cant speedup that parallel computing can provide. Since the primary reason for writing parallel codes is speed [13], it comes as no surprise that performance analysis is a vital part of the development process. Analysis tries to determine if a given algorithm is as fast as it can be, where the program can be further optimized, and how efficiently the underlying system is being used.

Department of Computer Science, Rensselaer Polytechnic Institute, Troy, NY

12/05/1998

cs-98-05